search_property_compsLABSCO SUMMARY
Sale prices that were filed with a registry, not modelled by someone.
Every figure it returns came from an official record — the Land Registry, the Dubai Land Department, a county assessor — so a comp check tells you what a property sold for rather than what a model guesses it is worth. Coverage is by market, and the 16 are listed up front.
What it is
Comparable-sales lookup over filed transaction records — HM Land Registry, DVF, the Dubai Land Department, US county assessors — covering 16 markets.
What you get
- Comparable sales near a postcode or ZIP, drawn from recorded transactions rather than listings or estimates
- Area statistics such as the median price for a named neighbourhood
- The market list, with the registry each one is sourced from
- Prices in each market's own currency — GBP, EUR, AED, SGD, TWD, USD
- The whole registry behind each market rather than a recent slice, with the UK the deepest of them
Requirements
Any MCP-compatible client — Claude Desktop and Claude Code are the named ones. Records are served from New Way Capital Advisory's API at api.nwc-advisory.com, which is also published on RapidAPI.
Setup effort
One command — pip install mcp httpx
USE CASES
Checking what a flat is worth before you offer
Ask for recent sales near the postcode and read the actual transaction prices, dates and property types.
Getting a read on a neighbourhood you do not know
Median prices for the area answer the question an agent's brochure will not.
Comparing two cities
Both markets have to be covered before the comparison means anything — check the market list, then pull each one's area statistics.
Finding out whether your area is covered at all
The market list names each country or city and the registry behind it, so a gap shows up before you build a question on it.
Backing a valuation with evidence
Pair the individual comparable sales with the area median so the number has both a sample and a baseline behind it.
3 tools
Listed in project docs. Your client may show a different list.
get_area_statslist_marketsBEFORE YOU INSTALL
What to know first
Tools you get
3 tools named in the project's documentation.
Requires
python on PATH
INSTALLATION
Connect your client
Server setup
pip install mcp httpxMCP Clients
Claude Desktop
{
"mcpServers": {
"property-comps": {
"command": "python",
"args": [
"<YOUR_PATH_TO_MCP_SERVER>"
]
}
}
}Claude Code
claude mcp add property-comps python <YOUR_PATH_TO_MCP_SERVER>